Putting Knowledge in Context to Facilitate Collaboration

摘要

Today's knowledge workers frequently suffer from problems associated with information overload, changing missions, and workforce turnover. The Contrail project is addressing how to improve personal and workgroup information management and collaboration in this environment. Contrail leverages machine-understandable models of users' contexts automatically built up by observing users performing information-keeping actions. We define context as those factors that bound and surround a problem. Contrail develops context models using text analytics, semantic networks, and spreading activation techniques to facilitate information retrieval and collaboration. The former leverages a knowledge base ("shoebox") of information items automatically maintained by Contrail. Items in this shoebox are tagged with their original context. This facilitates information retrieval, recall, and sharing among collaborators. Contrail has been focused on aiding intelligence analysts but is generally applicable to other knowledge workers. This paper describes the motivation for this work in context-aware computing and ethnographic studies of web usage; the Contrail tools; and a preliminary experiment to evaluate their utility for intelligence analysts.